At transmission of data files from a transmitting unit, or a server, to a receiving unit, for instance over Internet, it frequently occurs that the transmitting unit, or the network that is utilized for the transmission, is overloaded. This may result in that the transmission speed from a server to the receiving unit will be very low, which in its turn results in very long transfer times and also can result in that the transmission will be unsuccessful.
However, it is usual, i.a. on the Internet, that files are copied and distributed to a plurality of servers in order to share the load on these servers that files that are frequently requested to be transmitted can result in. This will reduce the risk of overloading and, of course, also opens for the possibility to break off a too slow transmission of a file, and request a transmission of the file from another server.
A problem in the situation is that the user or the receiving unit does not know which server, or which part of the network, that for the time being has a low load. By that there also will be the risk that also this server is, or will be, overloaded and that the transmission is slowed down or has to be broken. There also is a risk of imbalance of networks and servers.